Bulletin of the American Physical Society
71st Annual Meeting of the APS Division of Fluid Dynamics
Volume 63, Number 13
Sunday–Tuesday, November 18–20, 2018; Atlanta, Georgia
Session D21: Experimental Techniques: Volumetric PIV/ PTV |
Hide Abstracts |
Chair: Callum Gray, LaVision Room: Georgia World Congress Center B309 |
Sunday, November 18, 2018 2:30PM - 2:43PM |
D21.00001: 3D-IPTV: Single camera 3D Intensity Particle Tracking Velocimetry using particle intensities and structured light. Andres Alejandro Aguirre-Pablo, Abdulrahman B Aljedaani, Jinhui Xiong, Ramzi Idoughi, Wolfgang Heidrich, Sigurdur T Thoroddsen A single camera and a consumer grade LCD projector is used to achieve 3D Intensity Particle Tracking Velocimetry (3D-IPTV). This is possible using a structured monochromatic illumination sequence from the projector perpendicular to a camera. The 2D motion projection of the particles is recorded with the video camera. We encode the depth position of each particle (perpendicular to the image plane of the camera) in their intensity relative to their intrinsic brightness, to obtain full 3D paths. A 3D laser engraved glass cube with 1100 known spatial distribution defects is used for calibration. The calibration process includes the light-intensity field calibration and spatial distortion corrections from the projected light divergence and optical walls. Using a combination of changing light patterns and multiple sub-gradients of intensity, we can achieve 200 depth levels of resolution in a volume of approximately 60 x 60 x 50 mm3. |
Sunday, November 18, 2018 2:43PM - 2:56PM |
D21.00002: A novel particle tracking technique using a scanning laser setup tested via numerical experiment Melissa Kozul, Vipin Koothur, Nicholas Worth, James Dawson We propose a novel robust 3D particle tracking technique based on a scanning laser setup. The present aim is to measure both Eulerian and Lagrangian statistics in densely-seeded turbulent flows with good spatial and temporal resolution, seeking to overcome inherent difficulties with line-of-sight based volumetric methods. To do this we have developed an effective triangulation virtually eliminating ghost particle reconstruction using images from only two cameras. A laser sheet is rapidly traversed ('scanned') across a measurement volume illuminating only a thin slice of the flow at a time. The triangulation process is further improved by using a fitted sheet number for individual particles to fix their true location along the scan direction (Knutsen et al., Exp. Fluids (2017) 58:145) instead of assuming the nominal sheet position where they are imaged. Following successful reconstruction of a time series of 3D particle fields, Lagrangian velocities and accelerations are calculated using particle tracking. The method is verified via numerical experiment using a DNS database. The present technique reconstructs a high fraction of known synthetic particle locations for densities above 0.05ppp. Comparison of a variety of statistics is undertaken to test the fidelity of the method. |
Sunday, November 18, 2018 2:56PM - 3:09PM |
D21.00003: Fiber Optic Approach to Time Resolved Tomographic Particle Image Velocimetry Jonathan Eliel Reyes, Kareem Ahmed PIV has been heralded as an accurate means of extracting velocimetry information for the study of flow field phenomena. Many discoveries have been made utilizing the diagnostic, but the interpretation of the data is limited to two dimensions. In turbulent flows, three dimensionality is more often fact than exception. To accurately describe such flows, the PIV technique was expanded to 3D by use of tomography. The approach, though, suffers from lack of versatility, requires full optical accessibility, and is costly, limiting its use in laboratories. Presented are the results of a new approach to the tomographic PIV technique for acquiring time-resolved, three dimensional information to measure velocity and flame dynamics. The technology leverages fiber-optics and two high-speed cameras to capture eight image views of a domain to reconstruct and resolve a three-dimensional flow field. A camera sensor is split into four images by fiber couplings and the distal ends are arranged for 3D light-field capture. The key is the compactness of the system. This unique technology combines the capability of a multi-camera array without the complexity of a multi-camera arrangement, and sacrifice of resolution. |
Sunday, November 18, 2018 3:09PM - 3:22PM |
D21.00004: Development of a Modular, High-Speed Plenoptic Camera for Flow Tomography Zu Puayen Tan, Brian S Thurow Modern 3D flow tomography tools rely on simultaneous imaging of the test volume from multiple perspectives. For time-resolved measurements, this requires using one high-speed camera for each perspective, which can be cost-prohibitive. Furthermore, many experimental facilities (e.g., high-pressure rigs) do not have the optical access to allow for multiple cameras. To address this challenge, our study focused on the development of a Modular Plenoptic Adaptor (MPA) to enable light-field imaging capability in standard high-speed cameras. A single MPA-enabled camera is capable of time-resolved 3D PIV measurements, while 2-4 cameras may be used for 3D scalar-field measurements (compared to traditional tomography’s >7). In this talk, we introduce the principles of plenoptic flow measurement, and showcase key results from preliminary tests of the MPA, which include: (i) proof of spatially accurate (<1% error) reconstruction within a 50x32x50mm volume, (ii) post-capture refocusing and perspective manipulation in high-speed videos and (iii) volumetric reconstruction of a droplets field along with time-resolved 3D PIV measurements. Proposed future works include validating the MPA system’s PIV performance on canonical flow-fields such as a vortex ring and uniform flow.
|
Sunday, November 18, 2018 3:22PM - 3:35PM |
D21.00005: 3D particle location from perspective shifted plenoptic images Elise Hall, Daniel Guildenbecher, Brian S Thurow The application of plenoptic imaging to 3D analysis of particle field diagnostics is an emerging area of diagnostic development. A plenoptic camera uses a microlens array to collect the angular and spatial information of the incoming light rays, which can be manipulated in post processing to provide a 3D representation of a scene from a single snapshot. This work develops and tests an algorithm to determine 3D position by exploiting the perspective-shift capabilities of the plenoptic camera for measurement of explosively generated fragments. The algorithm is validated using a data set previously examined in a refocusing based study of a static particle field. Application to fluid dynamics measurements is demonstrated in examination of experimental data sets. The secondary droplet field created by the impact of a drop of water on a thin film of water and the fragment field of a lab scale detonator are measured to provide examples of particle size and shape variation. |
Sunday, November 18, 2018 3:35PM - 3:48PM |
D21.00006: Uncertainty Quantification in Volumetric Particle Tracking Velocimetry (PTV) Sayantan Bhattacharya, Pavlos Vlachos Volumetric PTV resolves 3D flow structures by tracking the motion of tracer particles seeded in a fluid. Recent advances in 3D PTV through iterative particle reconstruction methods like STB have shown increased reconstruction accuracy even for higher seeding densities. However, predicting the uncertainty in such a measurement is challenging due to the underdetermined system of equations associated with the inverse reconstruction problem, coupled with various factors affecting the calibration and tracking process. Here, we quantify the uncertainty in a particle based volumetric reconstruction and subsequently in a 3D PTV measurement. The reconstructed particle position uncertainty is a combination of the uncertainties in the volumetric calibration coefficients and the particle image position estimation. The LSQ fit uncertainty in the mapping function coefficients contributes to the calibration uncertainty. The uncertainty due to the mismatch between the projected and actual particle image locations is also considered. Finally, an uncertainty propagation through the 3D reconstruction equation gives the 3D position uncertainty, which directly affects the uncertainty in a 3D tracking process. The proposed methodology is tested for synthetic uniform flow and vortex ring cases. |
Sunday, November 18, 2018 3:48PM - 4:01PM |
D21.00007: Effect of density gradients on the Cramer-Rao lower bound for volumetric PIV/PTV measurements Lalit K Rajendran, Sayantan Bhattacharya, Sally PM Bane, Pavlos Vlachos Volumetric PIV/PTV experiments in variable density environments suffer from aero-optical effects due to refraction of light rays passing through the flow field. The density/refractive index gradients affect the Mie-scattered intensity, position and diameter of the projected particle image, and influence the accuracy and precision of the particle position estimates from the projected images. This effect varies with the viewing direction, particle position within the laser sheet, optical setup and density gradients in the flow field. We propose to theoretically study this effect by deriving the Cramer-Rao Lower Bound (CRLB) for the precision of the particle position estimation process. We model the image of a particle in the presence of density gradients, building on a previous particle image model developed for Background Oriented Schlieren (BOS) experiments, and compare theoretical predictions of the particle position estimation variance from the CRLB to ray tracing simulations. The results of this analysis can help elucidate the dependence of the position error on the experimental parameters, aid in experiment design, and provide directions in improving current reconstruction tools to account for these effects. |
Sunday, November 18, 2018 4:01PM - 4:14PM |
D21.00008: High-resolution particle-based 3D velocimetry using divergence-free radial basis functions Keishi Kumashiro, Adam Michael Steinberg, Masayuki Yano We present a new method of inferring high-resolution 3D divergence-free velocity fields from particle image tomograms. This method – termed tomographic particle flow velocimetry (T-PFV) – is based on representing the velocity field as a linear combination of divergence-free radial basis functions; the piece-wise constant representation of the estimated velocity field that is inherent to tomographic particle image velocimetry (T-PIV) is replaced by a smooth representation that automatically satisfies conservation of mass. The appropriate linear combination is determined using a non-regularized optical flow framework. We provide a detailed evaluation of T-PFV in terms of accuracy, spatial resolution, and sensitivity to parameters based on 3D constant-density DNS data. We also show that T-PFV yields substantial improvements in accuracy and spatial resolution compared to T-PIV over a wide range of parameters. |
Sunday, November 18, 2018 4:14PM - 4:27PM |
D21.00009: Dense Particle Identification and Reconstruction (DPIR) technique applied to simulated volumetric PTV images Dan Troolin, Aaron Boomsma, Wing T Lai DPIR is a three-dimensional particle tracking velocimetry technique that utilizes a three-pronged approach to identifying and reconstructing the positions and velocities of densely-packed particles within a volumetric domain for the purpose of understanding and analyzing fluid motion. The key aspects of the algorithm include a determination of the number of particles per support set prior to the final particle fitting; peaks, projections, and paths are used to determine the number of particles. These techniques are discussed in detail. DPIR is then applied to a set of simulated images with a known solution and compared to similar techniques. Results of the study include the yield of identified particles, the accuracy of particle positions, and the particle densities at which the results were realized. Further, the results will be discussed in the context of application to real images and future work. |
Sunday, November 18, 2018 4:27PM - 4:40PM |
D21.00010: DPIR: Dense Particle Identification and Reconstruction for Dual Frame and Time-Resolved Volumetric PTV Aaron Boomsma, Dan Troolin DPIR is a novel technique for improving the accuracy and reconstruction yield for both dual-frame and time-resolved volumetric PTV data. The technique utilizes peaks, projections, and paths within a recursive process to increase the accuracy of 2D particle identification. In high seeding densities, overlap ratios >50% are common, so DPIR fits multiple particle images within a support set using information from up to three sources: 1) image intensity peaks, 2) projections of previously triangulated particles, and 3) paths of 2D particle image trajectories in the image space. The technique is described and thoroughly assessed using synthetic particle images from both uniform and turbulent flows in low, medium, and high seeding densities. Notably, for the case of dual-frame data and ppp = 0.05, 95.3% of the real particles were reconstructed with 8.0% ghost particles in 15 iterations. The reconstruction error was 0.16 pixels. For time-resolved data, 2D paths were fit using a 2nd order polynomial for the case of turbulent flow with a high particle displacement of about 15 pixels. For this worst-case test, the paths returned image coordinate predictions with a maximum error of approximately 0.18 pixels at the ppp = 0.1, accurately fitting 93% of the particle images. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700